Proteolysis has been implicated not only in protein turnover, but also in the regulation of other physiological functions such as protein translocation, fibrinolysis, digestion, hormone maturation, and fertilization (see, e.g., Protein Degradation in Health and Disease in: CIBA Foundation Symposium 75, published by Excerpta Medica, N.Y. (1980)). Responsible for such functions are proteinases, which enzymatically catalyze the degradation of protein substrates.
A number of diverse mechanisms exist which play important roles in the regulation of proteolytic activity. Particularly important mechanisms include modulation of proteinase activity, modulation of substrates, localization of proteinases, and sequestration of proteinases into vesicles (see, e.g., Protein Degradation in Health and Disease in: CIBA Foundation Symposium 75, published by Excerpta Medica, N.Y. (1980)).
Calcium-activated neutral proteinase (CANP, or calpain), a non-lysosomal cysteine proteinase, is believed to participate in various intracellular processes mediated by calcium (Waxman, L., Methods Enzymol. 80:644-680 (1981); Imahori, K., In: Calcium and Cell Function, Vol. 3, pp. 473-487 (W. Y. Cheung, ed.), Academic Press, N.Y. (1982); Murachi, Trends Biochem. Sci. 8:167-169 (1983); Mellgren, R., FASEB J. 1:110-115 (1987); Schollmeyer, J. E., Science 240:911-913 (1988); Watanake et al. Nature 342:505-511 (1989)). CANPs comprise a family of enzymes that are widely distributed in mammalian and arian cells (Murachi et al., Trends in Biochem. Sciences 8:167-169 (1983)).
The potential activity of CANPs in cells appears to greatly exceed the physiological needs. For example, when maximally expressed under experimental conditions, CANPs in retinal ganglion cells can hydrolyze more than 50% of the entire content of axonal proteins within 5-15 minutes (Nixon et al., J. Neurosci. 6:1252-1263 (1986)). Given this enormous potential activity, it is not surprising that various mechanisms exist to regulate calcium-dependent proteolysis in vivo, including cellular calcium levels, modulation of the enzyme's sensitivity to calcium, autolytic activation, and its interaction with endogenous activating and inhibitory factors.
The role of CANPs in mediating cellular responses to the calcium signal is becoming widely accepted (Suzuki et al., FEBS Lett. 220:271-277 (1987)). The evidence linking calcium-activated proteolysis to key regulatory processes is of particular interest. The work of Pontremoli et al., PNAS USA 84:3604-3608 (1987), for example, has provided direct evidence in intact cells for a role of CANP in the modulation of responses to external stimuli mediated by protein kinase C.
CANPs have been implicated in neurobiological phenomena ranging from transmembrane signaling and synaptic plasticity to disease-related neuronal degeneration and cell death. This suspected functional diversity reflects the existence of multiple CANPs and a complex regulation which includes changes in calcium concentration, interaction with specific endogenous inhibitors (calpastatins), autolytic conversion of pro-forms to more active enzymatic forms, ability of the protease to translocate to different cellular sites, and influences of protein substrate properties.
The ability of CANPs to modulate protein kinase C and other protein kinases suggests that these proteinases are an important component of the regulatory system mediating the transduction of extracellular calcium signals. The ability of CANP to carry out limited proteolytic cleavage and thereby activate certain enzymes, including protein kinase C, emphasizes the potential creative functions of proteolysis and implies that the actions of CANP may be amplified intracellularly in provocative ways.
Because CANPs are particularly well-represented in neurons (Nixon et al., J. Neurochem. 6:1252-1263 (1986); Nixon, R. A., J. Neurochem. 6:1264-1271 (1986); Hamakabu et al., J. Neurosci. 6:3103-3111 (1986)), various neurobiological roles have been proposed for them, including synaptic modulation (Lynch et al., Science 224:1057-1063 (1984)), the post-translational modification and degradation of cytoskeletal proteins (reviews: Nixon, R. A., In: Neurofilaments (Marotta, C. A., eds.), pp. 117-154, Univ. Minnesota Press, Minneapolis, Minn. (1983); Schlaepfer, in: Neurofilaments, pp. 57-85, C. A. Marotta (ed.), Univ. of Minnesota Press (1983)) and the modification of membrane receptors (Baudry et al., Science 212:937-938 (1981)). CANPs appear to be involved in the postsynaptic events of long-term potentiation (Staubli et al., Brain Res. 444:153-158 (1988)), and the mounting evidence for a regulatory role of CANPs in protein kinase C modulation (Suzuki et al., FEBS Lett. 220:271-277 (1987)) is consistent with a role in presynaptic mechanisms of long-term potentiation (Akers et al., Science 231:587-589 (1986)).
Although the involvement of proteolytic systems in cell death is clearly established, there has been a traditional bias toward regarding proteolysis as an end-stage phenomenon or as merely a scavenging mechanism for cellular debris. This traditional role for proteolytic enzymes has been revised by a new appreciation of the varied regulatory roles of proteolytic enzymes in nearly every aspect of cellular function, including neurophysiological function.
Increased interest in the role of excitatoxins in mechanisms of cell death in neurodegenerative disease and related disorders (Spencer et al., Science 237:517-519 (1987)) also implicates CANPs. Rapid influx of calcium can lead directly to an activation of latent CANP. Some recent studies demonstrate that excitatoxins activate CANP and induce structural protein breakdown in vivo (Noszek, et al., Soc. Neurosci. Abstr. 13:1684 (1987)) and that CANP inhibitors reduced the damage produced by kainate, NMDA and quisqualate (Siman et al., Soc. Neurosci. Abstr. 13:1684 (1987)).
Such considerations are especially pertinent to an examination of Alzheimer's Disease pathogenesis, for example. It has been found that the inactivation of CANP with a specific monoclonal antibody inhibits the conversion of membrane-bound protein kinase C to a soluble calcium-independent form, thereby increasing the production of superoxides and stimulating the phosphorylation of membrane proteins. These effects are directly relevant to known mechanisms of cell death such as free radical-induced cellular damage, secondarily increased Ca.sup.2+ influx (Cross et al., Ann. Int. Med. 107:526-545 (1987)), and the structural abnormalities found in Alzheimer's Disease brain such as altered phosphorylation of cytoskeleton proteins (tau) and membrane damage leading to altered amyloid precursor protein processing (Dyrks et al., EMBO J. 7:949-957 (1988)).
The conspicuous susceptibility of cytoskeletal proteins to CANPs and the relative enrichment of CANPs in neurons has focused particular attention on these proteases as regulators of neuronal cytoskeleton dynamics.
By what mechanisms CANP activity may become down-regulated and thereby give rise to intracellular accumulations of cytoskeletal proteins or protein fragments in affected Alzheimer's Disease neurons, and at what stage calcium influx ultimately increases to activate latent CANPs to cause irreversible cell death, remain unanswered questions of considerable importance to Alzheimer's Disease and other late-onset neurodegenerative disorders. That is why the regulators of CANP activity, such as endogenous inhibitors, are so important.
Calpastatins, the specific protein inhibitors of CANP, are also widely distributed among tissues. First identified in 1978 (Waxman et al., J. Biol. Chem. 253:5888-5891 (1978)), calpastatins have since been purified from several different sources. Although each of the purified species shares the properties of heat stability and strict specificity for CANP, there is no consensus on the number of forms of calpastatin within single cells or among different cell types. The recent characterization of a calpastatin cDNA isolated from a rabbit cDNA library (Emori et al., Proc. Natl. Acad. Sci. USA 84:3590-3594 (1987)) revealed a deduced sequence of 718 amino acid residues (M.sub.r =76,964) containing four consecutive internal repeats of approximately 140 amino acid residues, each expressing inhibitory activity (Emori, et al., ibid. (1987)). This deduced molecular weight is significantly lower than the molecular weight of rabbit skeletal muscle calpastatin (M.sub.r =110,000), suggesting that the inhibitor migrates anomalously on SDS gels and may be post-translationally modified.
Other studies suggest that additional molecular forms of calpastatin may be present in tissues. Although 110 kDa calpastatin is observed in rabbit and bovine skeletal muscle (Nakamura et al., J. Biochem. 98:757-765 (1985); Otsuka et al., J. Biol. Chem. 262:5839-5851 (1987)), porcine cardiac muscle (Takano et al., J. Biochem. 235:97-102 (1986)) and human liver (Imajoh et al., FEBS Lett. 187:47-50 (1984)), other molecular forms of calpastatin have also been isolated, including a 68 kDa form from chick skeletal muscle (Ishiura et al., Biochem. Biophys. Acta 701:216-223 (1982)) and porcine erythrocytes (Takano et al., J. Biochem. 235:97-102 (1986)), a 50 kDa heterodimer from rabbit skeletal muscle (Nakamura et al., J. Biochem. 96:1399-1407 (1984)) and 34 kDa forms from rabbit skeletal muscle (Takahashi-Nakamura et al., J. Biochem. 90:1583-1589 (1981)) and rat liver (Yamato et al., Biochem. Biophys. Res. Comm. 115:715-721 (1983)). The sensitivity of calpastatin to proteolysis has suggested that smaller polypeptide chains containing inhibitory activity might be derived from larger precursors during purification, or in vivo. Although certain of these low molecular weight calpastatins resemble the higher molecular weight forms, their derivation from the same gene product has not been established.
Although the activity of calpastatins in the nervous system is considerable, little else is known about the properties of these proteins and how they regulate calcium-mediated proteolysis in neural cells.